AutomationBench-AA - Marketing - Raw Objective Completion: leaderboard
Artificial Analysis's implementation of Zapier's AutomationBench restricted to its Marketing tasks: the share of task objectives an agent completes in simulated SaaS applications, counted before the guardrail-violation adjustment used by the AutomationBench-AA headline. Reported 0-100 (%); higher is better.
Metric: Raw Objective Completion (%). Source: artificialanalysis.ai. Saturation forecast: Around February 2027. 243 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Gemini 4 Argon (High) | 94.66 | #9 (Gemini 4 Argon) |
| 2 | Gemini 4 Carbon (High) | 94.44 | #1 (Gemini 4 Carbon) |
| 3 | Kimi K3.1 Pro | 93.18 | #17 |
| 4 | MiMo-V2.6-Pro | 91.11 | #48 |
| 5 | Claude Opus 5 (Max) | 90.76 | #16 (Claude Opus 5) |
| 6 | MiMo-V2.6-Flash | 90.12 | #74 |
| 7 | Ling 3.1 Flash | 89.72 | #75 |
| 8 | Qwen3.8-Flash-Next | 89.1 | #76 |
| 9 | Qwen3.8 2.4T A95B | 88.86 | #59 |
| 10 | DeepSeek V4.1 Flash (1017) (High) | 88.79 | #27 (DeepSeek V4.1 Flash (1017)) |
| 11 | MiniMax-M3.1-Flash | 88.22 | #52 |
| 12 | Qwen 3.8 Max (0902) | 87.91 | #36 |
| 13 | Grok 4.6 (Medium) | 87.22 | #29 (Grok 4.6) |
| 14 | GLM-5.3 (Max) | 86.98 | #44 (GLM-5.3) |
| 15 | Claude Opus 5 (xHigh) | 86.96 | #16 (Claude Opus 5) |
No result here: #3 Claude Opus 5.5, #10 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=automationbench-aa-marketing-raw-objective-completion · How It Works · Data refreshed daily, snapshot 2026-11-05.